Review

Reviewed Work(s): Discovering Suicide: Studies in the Social Organisation of Sudden Death
by J. Maxwell Atkinson
Review by: Jason Ditton
Source: The British Journal of Criminology, Vol. 19, No. 2 (April 1979), pp. 188-190
Published by: Oxford University Press
Stable URL: https://www.jstor.org/stable/23636439
Accessed: 18-11-2018 19:35 UTC

JSTOR is a not-for-profit service that helps scholars, researchers, and students discover, use, and build upon a wide
range of content in a trusted digital archive. We use information technology and tools to increase productivity and
facilitate new forms of scholarship. For more information about JSTOR, please contact support@jstor.org.

Your use of the JSTOR archive indicates your acceptance of the Terms & Conditions of Use, available at
https://about.jstor.org/terms

Oxford University Press is collaborating with JSTOR to digitize, preserve and extend access
to The British Journal of Criminology

This content downloaded from 58.27.197.159 on Sun, 18 Nov 2018 19:35:47 UTC
All use subject to https://about.jstor.org/terms
 REVIEWS

showroom windows) may be surprised to note that 60 per cent, of th

against whom a compensation order was made had no more than £25
quarter had not completed payment within 18 months of sentence but o
cent, were committed to prison in default and of these nearly half w
serving a term of imprisonment so that the particular committal m
been a tidying up operation.
It is suggested that the scope for increasing the use of compensatio
greatest in relation to cases of wounding or assault. There are probl
lating to technicalities of assessment and the absence of familiar gu
these may lend themselves to solution more easily than those raised for
who attempts to obtain damages and costs from an individual thro
litigation. The criminal court has more effective means of enforcement.
for the victim whose injuries are likely to be assessed at under £150 and
fore not eligible for compensation by the Criminal Injuries Board the m
court could be much the best forum.
Mr. Softley has produced a most interesting report which is full of suggestive
points. His findings should certainly help to import a greater degree of reality into
discussions of reparation as a principle in sentencing.
W. E. Cavenagh

Discovering Suicide: Studies in the Social Organisation of Sudden Death.

By J. Maxwell Atkinson. [London: Macmillan Press. 1978. 225 pp.
£8-95-]

The only thing that Max Atkinson hasn't had to contend with in the last 10 years
(as far as I know) is all his notes and data catching fire. But he has already had
more than an average intellectual lifetime's dose of the remaining two academic
nightmares: first, being pipped to the post with the much needed critique of
Durkheim by that godfather of the publishing Mafia, Jack " The Editor "
Douglas; and secondly (and subsequently) undergoing massive open-minded
theoretical transplant surgery through his own personal conversion to ethno
methodology. Nevertheless, Atkinson does not allow such personal troubles and
changes to mar Discovering Suicide, which emerges as a wide-ranging, scholarly
and very readable text on all the sociology—and possible sociology—of suicide.
The solid first part of the book concentrates upon an extensive review of relevant
" suicidology " literature (a project distilled from his postgraduate studying under
Alasdair Maclntyre who had set him the task of tracing and reading everything
ever written on the subject!), which includes a diligent dismasting of positivism,
not a straw-man but still the tin-man of suicide studies, and the tin-God of most
sociology. From this carefully worked out critical backdrop, Atkinson confronts
central sociological resources like official statistics from a logically radical ethno
methodological position which takes us programmatically beyond both Sudnow,
and Kitsuse and Cicourel.
If there are minor criticisms, they are that Discovering Suicide is a little too
dependent on Atkinson's own previous suicide publications (as topic though,
rather than resource—perhaps there is something here for the sociologist of
sociology as well as the sociologist of suicide). Perhaps, also, the urge to " re-do "
188

This content downloaded from 58.27.197.159 on Sun, 18 Nov 2018 19:35:47 UTC
All use subject to https://about.jstor.org/terms
 REVIEWS

other work ethnomethodologically might have been resisted more con

The phrase " provide for " is used in its newer technical, conversationa
sense in the early part of the book, but this seems to elide with conventio
in the middle sections in (thus ambiguous) sentences like " provide sup
verdict of suicide " and "provide the basis for a decision " (p. 117, my e
I should also have thought that the sequential model of suicide process
have been constructed to include Becker's two paradoxical categories: th
accused, and those who got away with it. Although such inclusion w
unnecessary to provide a base for the analysis which follows it, they ar
gories upon which Atkinson's earlier and masterly critique of official stati
based.
Now, curiously, although Atkinson modestly hedges the presentation of his own
data with apologies like " the discovery of his [Sack's] works came too late for any
of the implications to be followed up in any detail " (p. 195) his analysis of the
methods used by British coroners in their production of suicides manages to be both
ethnomethodologically acceptable and specifically relevant to anybody's analysis
of suicide as a topic. Atkinson's early work stood out because he was prepared to
accept the labelling position absolutely and study the processes of control to the
exclusion of any lapse back into studying the subjects (a programme perhaps forced
on him by the procedural inability to do much interpretative work on dead bodies).
Discovering Suicide builds on the early work and describes in exhaustive detail what
features of those 20 per cent, of all mortalities which are referred to coroners are
called on by them to constitute both the explanans and (simultaneously and re
flexively) the explanadum. These are shown to be the technique of death (revealed
as suicide through the presence of " hesitation cuts " near great wounds, folded
clothes near drowned corpses or more than 10 barbiturates in the stomachs of
overdosed ones), the mode of death (of car deaths, those in smashed up vehicles
are held to be accidents, those in exhaust fume-filled ones, suicides), the place of
death (an overdose in bed being more accidental than one outside), other cir
cumstances (prescribed barbiturates being a more accidental form of death than
stolen ones), and biographic features (with depression being a commonly found
biographic feature of those deaths labelled suicidal). This sort of analysis allows
us to make sense not only of " obvious " cases, headlined as, for example,
" DEPRESSED " WIFE TOOK OVERDOSE OF DRUGS, but also of more
arcane examples like MAN WHO PLANNED TO VISIT AUSTRALIA
TOOK HIS OWN LIFE, and WOMAN TOOK OVERDOSE AFTER
WATCHING CHURCHILL'S FUNERAL.
There is a conventional tension in ethnomethodological analy
studying particular practical contents, and yet usually so in order
general methodological propositions. Critics who are blind to the
the programme make accusations of triviality or confuse the prog
symbolic interactionism. Often those blind to the first think th
associated with the second are conventional ones, with one " emin
sociologist " telling Atkinson " that the problem of indexicality c
circumstances be overcome by careful piloting of questionnaires
himself is nicely aware of the depth of the ethnomethodological
showing, for example, how lay theorising and conventional suicidology
189

This content downloaded from 58.27.197.159 on Sun, 18 Nov 2018 19:35:47 UTC
All use subject to https://about.jstor.org/terms
 REVIEWS

by a common methodic approach) but deals with suicide also as a top

sum, whereas most sociologists use Durkheim's Suicide as a resource meth
Atkinson's Discovering Suicide will be used both as a cogent resource
interested in method, and as a substantial and innovatory contribu
analysis of its topic.
Jason Ditton

Personality Characteristics of Violent Offenders and Suicidal Individuals.

By Liisa Jarvinen. [Helsinki: Suomalainen Tiedeakatemia. 1977. 195 pp.
statistical appendices].

This book, the eleventh in a series of published academic dissertations, does n

make easy reading. It has the disadvantages of a thesis that does everything
demonstrate the author's familiarity with a highly specialised literature, and
ability to deploy advanced statistical techniques, while doing little to help t
ignorant grasp the essentials of the problem from among a mass of technical deta
The empirical section is a study, utilising interviews and psychological tests, of
sample of 68 male prisoners serving sentences for homicide or aggravated assa
and a control group of 42 women brought to the Helsinki University Hospi
following suicide attempts. The purpose of the study was to test psychoanaly
hypotheses concerning the forms of personality disturbance that may underlie bo
homicidal and suicidal behaviour. The choice of women patients as a control group
for violent male criminals is too extraordinary to deserve further comment. It wa
dictated by the fact that it " proved difficult to obtain a control group consisting
entirely of men in the fixed time period " (p. 83).
The two projection tests from which the main findings were derived consisted o
the TAT (a series of pictures of ambiguous scenes about which the subjects have to
invent stories) and the Rorschach ink-blot test. Both of these involve the investigato
who scores them in complex and subjective judgments. The author herself, w
conducted the interviews with the prisoners, also seems to have scored the tests. I
these circumstances the inter-correlations between test scores and interview data,
however " significant " in a statistical sense, are not necessarily valid, and the applic
tion of factor analysis to the correlational matrix seems an irrelevant sophist
tion. Two questionnaires were also used. These presumably required no judgme
in scoring, but they yielded variables the reliability of which, " calculated by usin
the communalities of the factor analysis . . . remained so low that their use was pu
into question" (p. 173).
A few of the findings are readily understandable in commonplace criminological
terms. The 36 homicidal offenders, compared with the 32 assault offenders, more
often knew their victims and said they had been aggravated by them. They
expressed more suicidal thoughts and they were less conspicuous in their denial of
guilt feelings. The author does not discuss how far her results might be explicable
in terms of social situations rather than personal pathology. She finds, on the basis
of her projection tests, evidence for a degree of personality disorganisation " great
than behaviour during testing gave reason to assume " (p. 175). She concludes th
men guilty of grave violence " present problems which are primarily psychological
190

This content downloaded from 58.27.197.159 on Sun, 18 Nov 2018 19:35:47 UTC
All use subject to https://about.jstor.org/terms